Response integration in modular neural networks using Choquet Integral with Interval type 2 Sugeno measures

In this paper a new method for response integration, based on the Choquet Integral with Interval type-2 Sugeno measures is presented. The Choquet integral is used as a method to integrate the outputs of the modules of the modular neural networks (MNN). The fuzzy Sugeno measures of the Choquet integral are represented by an interval type-2 fuzzy system. A database of faces was used to perform the preprocessing, the training, and the combination of information sources of the MNN. Type-1 and interval type-2 fuzzy systems for edge detection based on the Sobel and Morphological gradient are used, which is a pre-processing applied to the training data for better performance in the MNN.